Subi Lee

I'm an undergraduate student at GIST. I'm currently working at Yonsei RLLAB advised by Prof. Youngwoon Lee. Previously, I focused on robotics learning in GIST AILAB advised by Prof. Kyoobin Lee.

Email  /  CV  /  Linkedin  /  Github

Research

My research focuses on reinforcement learning, with an emphasis on multi-task learning, unsupervised skill discovery, and robotic manipulation. I am particularly interested in improving generalization, sample efficiency, and the integration of large-scale vision-language models with control policies.

AMPED is a framework for skill-based reinforcement learning that simultaneously maximizes state coverage and skill diversity through several carefully designed components.

This research investigates Simplicial Normalization (SimNorm) as an activation function for multi-task reinforcement learning, showing that it underperforms ReLU in Meta-world benchmarks.

Website template from Jon Barron.